Over a million new cases of cancer will be diagnosed this year in the United States (see, for example, American Cancer Society; http://www.cancer.org/statistics/cff98/basicfact_toc.html; National Cancer Institute; http://rex.nci.nih.gov/NCI_Pub_Interface/raterisk/ratestoc.html). While surgery can often provide definitive treatment of cancer in its early stages, the eradication of metastases is crucial to the cure of more advanced disease. Chemotherapeutic drugs are used in combinations for this purpose, with considerable success. Nonetheless, over half a million Americans will die from cancer this year. Progressions and relapses following surgery and chemotherapy/radiation are not uncommon, and in most cases the second line of treatment is of limited use. Despite the expenditure of large amounts of public and private resources over many years, better treatments for cancer are sorely needed.
Currently there are approximately 100 antineoplastic drugs on the market. Their systemic use is associated with undesirable side effects including toxicity to normal cells, which limits the doses used for treatment of the disease. Most pharmaceuticals consist of small organic molecules, which effectively traverse cell membranes and become widely distributed through the body. As reviewed by Langer, polymer-based pharmaceutical agents provide a variety of new approaches to safer and better therapies (see, Langer R, Nature, 392 (6679) SUPPS: 5-10 (1998)). Polymers and other macromolecules do not traverse membranes; however, they may be selectively accumulated in the interstitial space of a tumor, since tumors typically do not possess an efficient lymphatic drainage system (Yuan et al., Cancer Research 51(12): 3119-30 (1991)). Developing technology to target therapeutic drugs to cancer cells, while sparing normal cells, is a promising approach to improved treatment; visualizing small cancers by means of targeting reagents is already a productive area of investigation.
The residence of macromolecules in tumors may be prolonged if they become anchored to immobile elements, such as polymorphic epithelial mucin (PEM), the secreted product of the MUC1 gene (Taylor-Papadimitriou et al., Trends Biotechnol., 12(6): 227-33 (1994)); or HLA-DR, a long-lived cell surface protein (Rose et al., Cancer Immunology Immunotherapy, 43: 26-30 (1996). The reagents of choice for this anchoring reaction are monoclonal antibodies and their derivatives. Currently there is a good selection of such macromolecules that bind to highly expressed tumor antigens, and that do not bind significantly to normal cells. Examples include, HMFG1 (Nicholson et al., Oncology Reports 5: 223-226 (1998)); L6 (DeNardo et al., Journal of Nuclear Medicine 39: 842-849 (1998)); and Lym-1 (DeNardo et al., Clinical Cancer Research, 3: 71-79 (1997)). The latter three antibodies have been conjugated to metal chelates for radioimmunotherapy and studied extensively in recent years, and are in clinical trials at various stages.
Recent data indicate that immunoconjugates have efficacy comparable to conventional antineoplastic drugs, and work in synergy with them (see, for example, Nicholson et al., Oncology Reports 5: 223-226 (1998); and DeNardo et al., Proceedings of the National Academy of Sciences USA 94: 4000-4004 (1997)). The emerging success of metal radioimmunoconjugates for cancer detection and treatment owes much to the development of metal-binding molecules (bifunctional chelating agents) appropriate for use in vivo, and to the further development of linkers that reduce concentrations of the metal binding molecules in nontarget tissues (see, Sundberg et al., Nature 250: 587-588 (1974); Yeh et al., Analytical Biochemistry 100: 152-159 (1979); Moi et al., Analytical Biochemistry 148: 249-253 (1985); Moi et al., Journal of the American Chemical Society 110: 6266-6267 (1988); and Li et al., Bioconjugate Chemistry 4: 275-283 (1993).
An alternative view of the potential for use of antibodies in cancer diagnosis and therapy is that, rather than carrying a radionuclide to a tumor, they can carry a receptor for a radionuclide (e.g., for a metal chelate). Antibodies against metal chelates can bind target molecules with high affinity and exquisite specificity and thus, can conveniently be used for applications in chemistry, environmental science, and medicine (see, e.g., Reardan et al., Nature 316: 265-268 (1985); Barbas et al., Proc. Natl. Acad. Sci. U.S.A. 90: 6385-6389 (1993); Blake et al., J. Biol. Chem. 271: 27677-27685 (1996); Brummer et al., Bioorg. Med. Chem. 9: 2253-2257 (2001); Schuhmacher et al., Cancer Res. 61: 3712-3717 (2001); Johnson et al., Environ. Sci. Technol. 36: 1042-1047 (2002)). For example, an antibody that binds rare earth complexes selectively could be used as a docking station for a set of probe molecules, of particular interest for medical imaging and therapy (see, e.g., Lubic et al., J. Nucl. Med. 42: 670-678 (2001) and Bosslet et al., Br. J. Cancer 63: 681-686 (1991)). The rare earths are rich in probe properties, such as the paramagnetism of Gd, the luminescence of Tb and Eu, and the nuclear properties of Y and Lu. The chelating ligand DOTA (1,4,7,10-tetraazacyclododecane-N,N′,N″,N′″-tetracetic acid) is generally useful for binding transition metals and rare earths with extreme stability under physiological conditions, leading to their use in vivo for targeted imaging and therapy (see, e.g., Loncin et al., Inorg. Chem. 25: 2646-2648 (1986); Geraldes et al., Magn. Reson. Med. 3: 242-250 (1986); Wu et al., Proc. Natl. Acad. Sci. U.S.A. 97: 8495-8500 (2000); and Waldherr et al., J. Nucl. Med. 43: 610-616 (2002)).
Antibodies with dual binding specificity have been prepared which can, e.g., cross-link tumor cells to cytokines such as tumor necrosis factor (Bruno et al., Cancer Res. 56(20): 4758-4765 (1996)). Likewise, bispecific antibodies that can bind to tumors and to metal chelates have been developed (Stickney et al., Cancer Res. 51(24): 6650-5 (1991); Rouvier et al., Horm. Res. 47(4-6): 163-167 (1997)). When pretargeted to tumors, these bispecific antibodies bind to antigens and remain on the target, providing receptors for metal chelates. Subsequent administration of small, hydrophilic metal chelates leads to their capture by the targeted chelate receptors. Uncaptured chelates clear quickly through the kidneys and out of the body, leaving very little radioactivity in normal tissues. This strategy is known as “pretargeting.”
A triumph of this approach was the imaging of metastatic cancer in the liver by an indium-111 chelate (Stickney et al., Cancer Res. B(24): 6650-5 (1991)). Antibodies conventionally conjugated to metal chelates are catabolized in the liver, and generally produce a radioactive background that masks tumors in that organ. The excellent tumor-to-background uptake ratios achieved by the pretargeting approach have led to considerable exploration of improvements in methodology. The anti-chelate antibody CHA255, initially developed for this purpose, possesses a high binding constant for (S)-benzyl-EDTA-indium chelates (Ks≈4×109) and exquisite specificity for these haptens (Dayton et al., Nature 316: 265-268 (1985). On CHA255, the bound lifetimes of various indium chelates at 37° C. were found to be in the 10-40 min range (Meyer, et al, Bioconjugate Chem. 1(4): 278-84 (1990)).
Another anti-chelate antibody, 2D12.5, possesses a high binding constant for (S) nitrobenzyl-DOTA chelates and Janus-DOTA, a bivalent form of DOTA (Ks≈108) (Goodwin et al., Canc. Res. 54(22): 5937-46 (1994) and Lubic et al., J. Nucl. Med. 42(4): 670-78 (2001)). In vivo clearance studies have demonstrated that 74-96% of 2D12.5 bound to metal chelates is present 24 hours after administration of the bound complex.
The need to enhance the antibody-hapten bound lifetime has led to the use of the long-lived avidin-biotin interaction, employing biotinylated metal chelates (Chinol et al., Nuclear Medicine Communications 18: 176-182 (1997)) in place of the original antibody-hapten interaction between CHA255 and benzyl-EDTA-indium derivatives. Here one assembles an antibody-avidin-chelate complex at the target in two or three steps, by sequential administration of nonradiolabeled proteins with a final administration of a biotinyl chelate carrying a radiometal. The extremely high affinity biotin-avidin association is adequately long-lived even for therapeutic applications (Theodore L J. et al, WO 9515979). Hen egg avidin and bacterial streptavidin, however, are both nonhuman, tetrameric proteins: their immunogenic properties are inconvenient, and the reversible associations between their subunits may limit their effectiveness. Thus, an improved strategy is still needed.
A delivery strategy based on the formation of a covalent bond between a chelate and an antibody that specifically recognizes and binds the chelate would represent a significant improvement over the methods now in use. The present invention provides engineered antibodies and chelates that react with one another to form covalent bonds and methods of using the engineered constructs to perform analyses and treat diseases.